Bridge hanging basket posture data analysis method and system

Abstract

The present application relates to the technical field of bridge construction, in particular to a bridge hanging basket posture data analysis method and system. A bridge hanging basket posture data analysis system comprises a disturbance action instruction set construction module, a connection state analysis module and an abnormality analysis module. The present application constructs posture-limited and linearly independent disturbance actions, obtains force-displacement responses of key connection points under the condition of not affecting the normal construction posture of the hanging basket, extracts equivalent stiffness, equivalent connection gap and hysteresis parameters, and realizes abnormality identification combined with abnormal type change trajectories, so as to effectively distinguish similar static responses caused by construction load changes and connection performance degradation.

Description

Technical Field

[0001] This invention relates to the field of bridge construction technology, specifically to a method and system for analyzing the posture data of a bridge hanging basket. Background Technology

[0002] Bridge formwork is an important temporary structure used to support formwork, construction loads, and newly poured beam segments during cantilever construction. It typically consists of a main truss, front and rear crossbeams, suspenders or straps, a rear anchor system, a traveling track, a base basket, a formwork system, and a hydraulic actuator. These components are connected by pins, bolts, anchor bolts, node plates, support contact surfaces, or welded structures, forming multiple stress transmission connection points. During repeated pouring, tensioning, unloading, and forward movement, the critical connection points of the bridge formwork are subjected to alternating tensile, compressive, shear, and local contact forces over a long period. This can easily lead to degradation of connection performance, such as reduced preload, widened gaps between pins and pin holes, detachment of support points, loosening of connections, decreased friction capacity, localized micro-slippage, and irreversible deformation. If these degradations are not identified in time, they may affect the overall stress stability and construction safety of the formwork.

[0003] Current bridge formwork condition monitoring mainly employs methods such as manual inspection, stress or displacement monitoring of key components, comparison of design values ​​with measured values, and verification of finite element calculation results. Manual inspection can detect obvious deformation, component damage, or loose exposed connectors, but its ability to identify hidden degradation states such as internal gaps in connections, loss of preload, changes in contact state, and early micro-slippage is limited. Detection methods based on stress, force, or displacement monitoring typically compare the collected data with preset thresholds and issue warnings when the monitored data exceeds the thresholds. However, the static force and displacement response of bridge formwork is simultaneously affected by factors such as concrete pouring volume, formwork and equipment location, distribution of construction personnel and materials, prestressing construction, wind load, temperature changes, and changes in the formwork's own posture. Changes in construction load and degradation of connection performance can both cause local increases in force, displacement, or redistribution of force, resulting in similar static responses. Simply relying on the force value, displacement value, or the deviation of the two from the design value at a certain moment makes it difficult to determine whether the response change originates from normal construction load changes or the degradation of the performance of the key connection points themselves. Summary of the Invention

[0004] This invention constructs an attitude sensitivity matrix and extracts its null space to generate multiple linearly independent perturbation action command sets without causing or only causing changes in the overall attitude of the hanging basket within permissible limits. These perturbation actions are then sequentially applied to actuators such as hydraulic jacks, traveling hydraulic cylinders, and adjustable booms to actively stimulate the local force-displacement responses of different key connection points under multiple independent perturbation directions. Based on the force response data and local relative displacement response data at the same moment, equivalent stiffness, equivalent connection gap, and hysteresis parameters are determined. After unified dimensional transformation, standard connection state parameters characterizing the connection's force transmission capacity, slackness, and energy consumption characteristics are formed. Simultaneously, physical degradation variables such as preload loss rate, gap increment, or friction coefficient loss rate are used as independent variables. By constructing continuous anomaly type change trajectories using smooth spline functions with non-negativity, range of values, and zero degradation state constraints, and identifying anomaly types based on the joint projection error of the current standard connection state parameters to each anomaly type change trajectory and the corresponding physical degradation variables, the system can separate the response amplitude changes mainly caused by construction load changes from the equivalent stiffness reduction, equivalent connection gap increase, and hysteresis characteristic changes caused by connection performance degradation. This overcomes the problem that construction load changes and connection degradation may produce similar responses and be difficult to distinguish when relying solely on the static displacement or force amplitude of the hanging basket. Ultimately, it enables the identification of the connection performance degradation type and its physical degradation degree at key connection points without interfering with the normal construction posture of the hanging basket or applying additional dangerous loads.

[0005] This invention also provides a method for analyzing the attitude data of bridge hanging baskets, including: Several disturbance action instruction sets are constructed based on linear independence constraints, and all disturbance action instruction sets are applied to the bridge hanging basket in sequence. Response sequence datasets corresponding to all key connection points of the bridge hanging basket are collected. The response sequence datasets include force response sequence data and displacement response sequence data of key connection points. The force response sequence data and displacement response sequence data of key connection points respectively include force response data and displacement response data of key connection points arranged according to the sampling time. Each response sequence dataset is associated with the type of a key connection point. Based on the correlation between the force response sequence data and the displacement response sequence data of the key connection points in the response sequence dataset, the connection state parameter dataset corresponding to each key connection point is determined. The connection state parameter dataset stores the connection state parameters, which include equivalent stiffness, equivalent connection gap and hysteresis parameters. For each critical connection point, after performing a uniform dimension transformation operation on the connection state parameter dataset corresponding to the critical connection point, a standard connection state parameter dataset is obtained. An anomaly type change trajectory library corresponding to the type of the critical connection point is obtained. The anomaly type change trajectory library includes the connection state parameter change trajectory corresponding to all connection state parameters for each anomaly type. The anomaly type corresponding to the critical connection point is determined based on the distribution of the projection of the standard connection state parameter dataset onto the connection state parameter change trajectory.

[0006] Preferably, several perturbation action instruction sets are constructed based on linear independence constraints, specifically including the following steps: Obtain the attitude sensitivity matrix, and denote the difference between the number of columns of the attitude sensitivity matrix and the numerical rank of the attitude sensitivity matrix as the null space dimension H. Perform complete singular value decomposition on the attitude sensitivity matrix, and then extract the last H column vectors from the right singular matrix in the output of the singular value decomposition to form the null space basis matrix. When it is necessary to generate K linearly independent candidate perturbation action vectors, an initial coefficient matrix of size H×K is randomly generated. The coefficients in the initial coefficient matrix are random numbers within a set range. The column vectors of the initial coefficient matrix are orthogonalized to obtain a coefficient matrix with full column rank. The null basis matrix is ​​multiplied by the coefficient matrix to obtain a candidate perturbation action matrix of size m×K, where K is not greater than q. Each column in the candidate perturbation action matrix corresponds to a candidate perturbation action vector. The attitude sensitivity matrix is ​​constructed as follows: several bridge cantilever formwork attitude analysis samples are obtained, including disturbance motion vectors and corresponding bridge cantilever formwork attitude vectors. The disturbance motion vectors include control quantities corresponding to several actuators. The attitude analysis model is constructed by multiplying the attitude sensitivity matrix and the transpose of the disturbance motion vectors, which equals the transpose of the bridge cantilever formwork attitude vectors. Then, all bridge cantilever formwork attitude analysis samples are substituted into the attitude analysis model, and a regression operation is performed on the attitude sensitivity matrix.

[0007] Preferably, the connection state parameter dataset corresponding to each key connection point is determined based on the correlation between the force response sequence data and the displacement response sequence data of the key connection points in the response sequence dataset, specifically including the following steps: The force response data and displacement response data of the key connection points at the same sampling time are time-aligned, and the force response data and displacement response data at the first sampling time in the force response sequence data and displacement response sequence data of the key connection points are subtracted respectively to obtain the force increment data and the local relative displacement increment data; the force increment data and the local relative displacement increment data at the same time are combined into response data points according to the sampling time order. Using local relative displacement increment data as independent variables and force increment data as dependent variables, linear regression is performed on all response data points to construct a response linear relationship. The slope in the response linear relationship is determined as the equivalent stiffness corresponding to the key connection point. The local relative displacement increment data corresponding to the force increment data of 0 in the response linear relationship is determined as the equivalent connection gap corresponding to the key connection point. The trajectory area formed between adjacent response data points is calculated according to the sampling time sequence, and the trajectory area is normalized by the variation range of the force increment data and the variation range of the local relative displacement increment data to obtain the hysteresis parameter.

[0008] Preferably, the anomaly type change trajectory library is constructed through the following steps: For any critical connection point, determine the physical degradation variable corresponding to the type of the anomaly, obtain several anomaly analysis samples, including the physical degradation variable and the corresponding standard connection state parameter dataset, and combine all the anomaly analysis samples into an anomaly analysis sample set corresponding to the anomaly type of the critical connection point. For any critical join point, the connection state parameters in the connection state parameter dataset are traversed, and the currently selected connection state parameter is determined as the target connection state parameter. The physical degradation variables and target connection state parameters contained in each anomaly analysis sample are extracted from the anomaly analysis sample set corresponding to the anomaly type. These two are combined to form a fitting point set, with the physical degradation variables as independent variables and the target connection state parameters as dependent variables. The sample fitting error is the sum of the squares of the differences between the predicted values ​​of the intermediate fitting function at each physical degradation variable and the corresponding target connection state parameter. The second derivative of the intermediate fitting function within the interval corresponding to the physical degradation variables covered by all anomaly analysis samples in the anomaly analysis sample set is used as the fitting error. The integral is used to obtain a smoothing penalty term. The sum of the product of the sample fitting error, the smoothing penalty term, and the smoothing coefficient is then used as the fitting objective. By adjusting the coefficients of the spline function in the intermediate fitting function, the fitting objective is minimized, resulting in the trajectory of the connection state parameters corresponding to the target connection state parameters. Each trajectory of the connection state parameters is a continuous function between the physical degradation variable and the corresponding connection state parameter. This process continues until all connection state parameters in the connection state parameter dataset have been traversed. All the trajectory of the connection state parameters are then combined into a set of anomaly type trajectory corresponding to the anomaly type. Finally, the sets of anomaly type trajectory corresponding to all anomaly types under the type of the key connection point are combined into an anomaly type trajectory library.

[0009] Preferably, the anomaly type corresponding to the key connection point is determined based on the distribution of the projection of the standard connection state parameter dataset onto the connection state parameter change trajectory, specifically including the following steps: Obtain the anomaly type change trajectory library, traverse all anomaly types, and for the selected anomaly type, record the square of the difference between the standard connection state parameter and the connection state parameter change trajectory corresponding to the target connection state parameter corresponding to the standard connection state parameter as the connection state parameter difference function. Then, perform a weighted summation operation on all connection state parameter difference functions to obtain the joint interpretation error function. The anomaly type corresponding to the critical join point is output if the optimal matching value is less than the matching threshold and the physical degradation variable corresponding to the optimal matching value is higher than the degradation threshold. If there is no anomaly type where the optimal matching value is less than the matching threshold and the physical degradation variable corresponding to the optimal matching value is higher than the degradation threshold, then the critical join point does not have anomalies.

[0010] Preferably, after performing a uniform dimensional transformation operation on the connection state parameter dataset corresponding to the key connection points, a standard connection state parameter dataset is obtained, which specifically includes the following steps: For the connection state parameters in the connection state parameter dataset, the ratio of the connection state parameter to the corresponding baseline quantity is recorded as the standard connection state parameter. Then, all the standard connection state parameters are combined into a standard connection state parameter dataset.

[0011] The present invention also provides a bridge hanging basket attitude data analysis system, comprising: The disturbance action instruction set construction module is used to construct several disturbance action instruction sets based on linearly independent constraints, and apply all disturbance action instruction sets to the bridge hanging basket in sequence. It also collects response sequence datasets corresponding to all key connection points of the bridge hanging basket. The response sequence datasets include key connection point force response sequence data and key connection point displacement response sequence data. The key connection point force response sequence data and key connection point displacement response sequence data respectively include key connection point force response data and key connection point displacement response data arranged according to the sampling time. Each response sequence dataset is associated with the type of a key connection point. The connection state analysis module is used to determine the connection state parameter dataset corresponding to each key connection point based on the correlation between the force response sequence data and the displacement response sequence data of the key connection points in the response sequence dataset. The connection state parameter dataset stores the connection state parameters, which include equivalent stiffness, equivalent connection gap and hysteresis parameters. The anomaly analysis module is used to perform a uniform dimension transformation operation on the connection state parameter dataset corresponding to each key connection point to obtain a standard connection state parameter dataset. It then obtains a library of anomaly type change trajectories corresponding to the type of the key connection point. The library of anomaly type change trajectories includes the connection state parameter change trajectories corresponding to all connection state parameters for each anomaly type. Based on the distribution of the projection of the standard connection state parameter dataset onto the connection state parameter change trajectories, the anomaly type corresponding to the key connection point is determined.

[0012] The present invention has the following advantages: This invention constructs an attitude sensitivity matrix and extracts its null space to generate multiple linearly independent perturbation action command sets without causing or only causing changes in the overall attitude of the hanging basket within permissible limits. These perturbation actions are then sequentially applied to actuators such as hydraulic jacks, traveling hydraulic cylinders, and adjustable booms to actively stimulate the local force-displacement responses of different key connection points under multiple independent perturbation directions. Based on the force response data and local relative displacement response data at the same moment, equivalent stiffness, equivalent connection gap, and hysteresis parameters are determined. After unified dimensional transformation, standard connection state parameters characterizing the connection's force transmission capacity, slackness, and energy consumption characteristics are formed. Simultaneously, physical degradation variables such as preload loss rate, gap increment, or friction coefficient loss rate are used as independent variables. By constructing continuous anomaly type change trajectories using smooth spline functions with non-negativity, range of values, and zero degradation state constraints, and identifying anomaly types based on the joint projection error of the current standard connection state parameters to each anomaly type change trajectory and the corresponding physical degradation variables, the system can separate the response amplitude changes mainly caused by construction load changes from the equivalent stiffness reduction, equivalent connection gap increase, and hysteresis characteristic changes caused by connection performance degradation. This overcomes the problem that construction load changes and connection degradation may produce similar responses and be difficult to distinguish when relying solely on the static displacement or force amplitude of the hanging basket. Ultimately, it enables the identification of the connection performance degradation type and its physical degradation degree at key connection points without interfering with the normal construction posture of the hanging basket or applying additional dangerous loads. Attached Figure Description

[0013] Figure 1 This is a schematic diagram of the bridge hanging basket attitude data analysis system used in an embodiment of the present invention. Detailed Implementation

[0014] To enable those skilled in the art to better understand the technical solutions of this invention, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of this invention.

[0015] Example 1: A method for analyzing the attitude data of a bridge hanging basket, comprising: Several disturbance action command sets are constructed based on linear independence constraints. These sets are then applied sequentially to the bridge formwork, and response sequence datasets are collected for all key connection points of the formwork. Each response sequence dataset includes force response sequence data and displacement response sequence data for key connection points, arranged according to sampling time. Each key connection point is associated with a specific type. Key connection points can be connections such as suspenders, anchors, the connection between the traveling track and the traveling device, or main truss member nodes on the bridge formwork. Operators can set up corresponding sensors at these locations to collect force and displacement response sequence data. The type of a key connection point is bound to the corresponding sensor, allowing identification of the key connection point type based on the data uploaded by the sensors, such as suspender or sling connection. Connection points include: the connection between the hanger and the upper or lower crossbeam or the base basket; the rear anchor connection point, which is the connection between the rear anchor rod, anchor beam, anchor plate and the cast-in-place beam segment; the support connection point, including the contact connection between the front and rear supports and the main truss, track, or beam; the travel system connection point, including the connection between the travel track joint, track clamping device, slipper or roller and the track; the main truss node, including the pin, bolt, weld, or node plate connection between the main truss members; and the crossbeam connection point, including the connection between the front and rear upper crossbeams and the main truss, hanger, and... Connections between baskets; connection points between baskets and templates, including connections between basket longitudinal beams, transverse beams, outer mold, inner mold and adjustment devices; hydraulic actuator connection points, including connections between the two ends of the jack and the main truss, support points, rails or adjustment components. The type of these key connection points also needs to be bound to the specific connection model and linked to specific material and size information to further distinguish the type of key connection points. For example, the type of key connection points corresponding to Q345 steel, M30 anchor rods, and rear anchor models with specific anchor plate sizes. Based on the correlation between the force response sequence data and displacement response sequence data of key connection points in the response sequence dataset, a connection state parameter dataset corresponding to each key connection point is determined. The connection state parameter dataset stores connection state parameters, including equivalent stiffness, equivalent connection gap, and hysteresis parameter. The equivalent stiffness represents the ability of the key connection point to resist local deformation after establishing effective force. The equivalent connection gap represents the idle travel or equivalent relaxation amount between the start of relative motion and the establishment of stable force transmission at the key connection point, and is used to reflect pin-hole clearance, fulcrum separation, connection slack, or insufficient preload. The hysteresis parameter represents the degree of non-overlap between the loading and unloading paths of the key connection point, and is used to reflect frictional energy dissipation, gap closure, local micro-slippage, and irreversible deformation. For each critical connection point, a standardized connection state parameter dataset is obtained by performing a unified dimension transformation on the connection state parameter dataset corresponding to the critical connection point. An anomaly type change trajectory library is then acquired, containing the connection state parameter change trajectories for all connection state parameters corresponding to each anomaly type. The anomaly type corresponding to the critical connection point is determined based on the distribution of the projections from the standard connection state parameter dataset to the connection state parameter change trajectories. Here, the connection state parameters are the change trajectories corresponding to equivalent stiffness, equivalent connection gap, hysteresis parameter, and slip threshold. The anomaly type is set by the operator, such as reduced preload, increased equivalent connection gap, and decreased friction coefficient. Several perturbation action instruction sets are constructed based on linear independence constraints, specifically including the following steps: Obtain the attitude sensitivity matrix. The difference between the number of columns of the attitude sensitivity matrix and the numerical rank of the attitude sensitivity matrix is ​​denoted as the null space dimension H. Perform complete singular value decomposition on the attitude sensitivity matrix. Then, extract the last H column vectors from the right singular matrix in the output of the singular value decomposition to form the null space basis matrix. Each column vector in the null space basis matrix represents a basic motion direction that has no effect or only a small effect on the overall attitude of the bridge hanging basket. Different column vectors of the null space basis matrix N are mutually orthogonal and linearly independent. When it is necessary to generate K linearly independent candidate perturbation action vectors, an initial coefficient matrix of size H×K is randomly generated. The coefficients in the initial coefficient matrix are random numbers within a set range. The column vectors of the initial coefficient matrix are orthogonalized to obtain a coefficient matrix with full column rank. The null basis matrix is ​​multiplied by the coefficient matrix to obtain a candidate perturbation action matrix of size m×K, where K is not greater than q. Each column in the candidate perturbation action matrix corresponds to a candidate perturbation action vector. The attitude sensitivity matrix is ​​constructed as follows: Several bridge cantilever basket attitude analysis samples are obtained. These samples include disturbance motion vectors and corresponding bridge cantilever basket attitude vectors. The disturbance motion vectors include control quantities corresponding to several actuators. The bridge cantilever basket attitude vectors include the overall displacement change of the bridge cantilever basket reference rigid body along the longitudinal direction of the bridge, the overall displacement change of the bridge cantilever basket reference rigid body along the transverse direction of the bridge, the overall displacement change of the bridge cantilever basket reference rigid body along the vertical direction, the transverse tilt angle change of the bridge cantilever basket reference rigid body, the longitudinal tilt angle change of the bridge cantilever basket reference rigid body, and the rotation angle change of the bridge cantilever basket reference rigid body in the horizontal plane. Operators can perform finite element analysis on the bridge cantilever basket. After setting the actuators according to the disturbance motion vectors, the corresponding bridge cantilever basket attitude vectors are collected, and bridge cantilever basket attitude analysis samples are constructed. The product of the attitude sensitivity matrix and the transpose of the disturbance motion vector is equal to the transpose of the bridge cantilever basket attitude vector. An attitude analysis model is constructed, and then all bridge cantilever basket attitude analysis samples are substituted into the attitude analysis model. A regression operation is performed on the attitude sensitivity matrix, which can be done using the least squares method. The connection state parameter dataset corresponding to each key connection point is determined based on the correlation between the force response sequence data and the displacement response sequence data of key connection points in the response sequence dataset. The specific steps include the following: The force response data and displacement response data of the key connection points at the same sampling time are time-aligned, and the force response data and displacement response data at the first sampling time in the force response sequence data and displacement response sequence data of the key connection points are subtracted respectively to obtain the force increment data and the local relative displacement increment data; the force increment data and the local relative displacement increment data at the same time are combined into response data points according to the sampling time order. Using local relative displacement increment data as independent variables and force increment data as dependent variables, linear regression is performed on all response data points to construct a response linear relationship. The slope in the response linear relationship is determined as the equivalent stiffness corresponding to the key connection point. The local relative displacement increment data corresponding to the force increment data of 0 in the response linear relationship is determined as the equivalent connection gap corresponding to the key connection point. The trajectory area formed between adjacent response data points is calculated according to the sampling time sequence, and the trajectory area is normalized by the range of change of force increment data and the range of change of local relative displacement increment data to obtain the hysteresis parameter. After performing a uniform dimensional transformation on the connection state parameter dataset corresponding to the key connection points, a standard connection state parameter dataset is obtained. The specific steps include the following: For the connection state parameters in the connection state parameter dataset, the ratio of the connection state parameter to the baseline quantity corresponding to the connection state parameter is recorded as the standard connection state parameter. The baseline quantity corresponding to the connection state parameter is the connection state parameter of the key connection point when no abnormality occurs. It is obtained by the operator in the calibration experiment. Then, all the standard connection state parameters are combined into a standard connection state parameter dataset. The anomaly type change trajectory library is constructed through the following steps: For any critical connection point, the corresponding anomaly type is determined, and the corresponding physical degradation variable is identified. For example, the preload loss rate is used as the physical degradation variable for the preload reduction anomaly, the equivalent connection gap increment is used as the physical degradation variable for the equivalent connection gap expansion anomaly, and the friction coefficient loss rate is used as the physical degradation variable for the friction capacity reduction anomaly. Several anomaly analysis samples are obtained, which include physical degradation variables and corresponding standard connection state parameter datasets. All anomaly analysis samples are then combined to form an anomaly analysis sample set corresponding to the anomaly type of the critical connection point. It should be noted that the anomaly analysis samples here are obtained by the operator through calibration experiments performed on the actual connectors corresponding to the connection point type. The operator sets different physical degradation variables for different anomaly types and then measures and constructs the corresponding standard connection state parameter datasets. For any critical join point, the connection state parameters in the connection state parameter dataset are traversed, and the currently selected connection state parameter is determined as the target connection state parameter. The physical degradation variables and target connection state parameters contained in each anomaly analysis sample are extracted from the anomaly analysis sample set corresponding to the anomaly type. These are combined to form a set of fitting points. Using the physical degradation variables as independent variables and the target connection state parameters as dependent variables, a cubic spline function is used to fit all fitting points, resulting in an intermediate fitting function. The cubic spline function leverages the local adjustment characteristics of spline functions, ensuring that the fitting adjustment within a certain interval of physical degradation variables primarily affects adjacent intervals, reducing the impact of local sample fluctuations on the overall trajectory of the connection state parameters. It also avoids the production of drastic oscillations without physical basis in higher-order polynomials at the ends of the fitting interval or between adjacent fitting points. The sample fitting error is the sum of the squares of the differences between the predicted values ​​of the intermediate fitting function at each physical degradation variable and the corresponding target connection state parameter. The sample fitting error measures the explanatory power of the intermediate fitting function for the anomaly analysis samples; a smaller sample fitting error indicates that the prediction results of the intermediate fitting function under each known physical degradation state are more accurate. The target connectivity parameters are used for smoothing. The integral of the second derivative of the intermediate fitting function over the intervals corresponding to the physical degradation variables covered by all anomaly analysis samples in the anomaly analysis sample set is used as the smoothing penalty term. The smoothing penalty term is used to suppress unnecessary bending caused by the intermediate fitting function to overfit individual anomaly analysis samples, but does not directly restrict the target connectivity parameters from increasing or decreasing normally with the physical degradation variables. The sum of the product of the sample fitting error, the smoothing penalty term, and the smoothing coefficient is used as the fitting target. The smoothing coefficient is set by the operator. By adjusting the coefficients of the spline function in the intermediate fitting function, the fitting target is minimized, and the connectivity parameter change trajectory corresponding to the target connectivity parameters is obtained. Each connectivity parameter change trajectory is a continuous function between the physical degradation variable and the corresponding connectivity parameter, representing the relationship between the target connectivity parameters and the physical degradation variable. This process continues until all connectivity parameters in the connectivity parameter dataset have been traversed. All connectivity parameter change trajectories are then combined into an anomaly type change trajectory set corresponding to the anomaly type. Finally, the anomaly type change trajectory sets corresponding to all anomaly types under the type of the key connectivity point are combined into an anomaly type change trajectory library. It should be added that physical constraints need to be added during the fitting process. For example, the parameters of the continuous function should be non-negative and the values ​​should not exceed the physical range defined by the anomaly analysis samples. When the anomaly analysis sample set includes zero-degenerate state samples, the intermediate fitting function should also be connected to the corresponding target connection state parameters at the zero-degenerate state. The anomaly type corresponding to the critical connection point is determined based on the distribution of the projection from the standard connection state parameter dataset to the connection state parameter change trajectory. The specific steps include the following: The system retrieves a library of anomaly type change trajectories, iterates through all anomaly types, and for the selected anomaly type, denotes the square of the difference between the standard connection state parameters and the connection state parameters corresponding to the target connection state parameters as the connection state parameter difference function. Then, it performs a weighted summation operation on all connection state parameter difference functions to obtain the joint interpretation error function. The weights for this weighted summation operation are set by the operator. The minimum value of the joint interpretation error function within its range is recorded as the optimal matching value corresponding to the selected anomaly type. The anomaly type corresponding to the critical connection point is output as the anomaly type when the optimal matching value is less than the matching threshold and the physical degradation variable corresponding to the optimal matching value is higher than the degradation threshold. Both the matching threshold and the degradation threshold are set by the operator. In particular, if there is no anomaly type where the optimal matching value is less than the matching threshold and the physical degradation variable corresponding to the optimal matching value is higher than the degradation threshold, then the critical connection point does not have anomalies. This application constructs an attitude sensitivity matrix and extracts its null space to generate multiple linearly independent disturbance action command sets under the condition of not causing or only causing changes in the overall attitude of the hanging basket within an allowable range. These disturbance actions are then applied sequentially to actuators such as hydraulic jacks, traveling hydraulic cylinders, and adjustable booms to actively stimulate the local force-displacement response of different key connection points under multiple independent disturbance directions. Based on the force response data and local relative displacement response data at the same moment, the equivalent stiffness, equivalent connection gap, and hysteresis parameters are determined. After unified dimension conversion, standard connection state parameters that can characterize the connection force transmission capacity, slack degree, and energy consumption characteristics are formed. At the same time, physical degradation variables such as preload loss rate, gap increment, or friction coefficient loss rate are used as independent variables. By constructing continuous anomaly type change trajectories using smooth spline functions with non-negativity, range of values, and zero degradation state constraints, and identifying anomaly types based on the joint projection error of the current standard connection state parameters to each anomaly type change trajectory and the corresponding physical degradation variables, the system can separate the response amplitude changes mainly caused by construction load changes from the equivalent stiffness reduction, equivalent connection gap increase, and hysteresis characteristic changes caused by connection performance degradation. This overcomes the problem that construction load changes and connection degradation may produce similar responses and be difficult to distinguish when relying solely on the static displacement or force amplitude of the hanging basket. Ultimately, it enables the identification of the connection performance degradation type and its physical degradation degree at key connection points without interfering with the normal construction posture of the hanging basket or applying additional dangerous loads.

[0016] Example 2: A bridge hanging basket attitude data analysis system, such as Figure 1 As shown, it includes: The disturbance action instruction set construction module is used to construct several disturbance action instruction sets based on linearly independent constraints. All disturbance action instruction sets are then applied sequentially to the bridge formwork, and response sequence datasets are collected for all key connection points corresponding to the bridge formwork. These response sequence datasets include force response sequence data and displacement response sequence data for key connection points. Each set contains force and displacement response data arranged according to sampling time, and each dataset is associated with a specific type of key connection point. Key connection points can be connections such as hangers, anchors, the connection between the traveling track and the traveling device, and main truss member nodes on the bridge formwork. Operators can set up corresponding sensors at these locations to collect force and displacement response sequence data. The type of key connection point is bound to the corresponding sensor, allowing identification of the key connection point type based on the data uploaded by the sensors. : Hoist or sling connection points, i.e., the connection between the hoist and the upper or lower crossbeam or the base basket; rear anchor connection points, i.e., the connection between the rear anchor rod, anchor beam, anchor plate and the cast-in-place beam segment; support connection points, including the contact connection between the front and rear supports and the main truss, track, or beam; travel system connection points, including the connection between the travel track joint, track clamping device, slipper or roller and the track; main truss nodes, including the pin, bolt, welding, or node plate connection between the main truss members; crossbeam connection points, including the connection between the front and rear upper crossbeams, lower crossbeams and the main truss. Connections between the hoists and the basket; connection points between the basket and the formwork, including connections between the basket's longitudinal beams, transverse beams, outer formwork, inner formwork, and adjustment devices; connection points of hydraulic actuators, including connections between the two ends of the jacks and the main truss, supports, tracks, or adjustment components. Furthermore, the type of these key connection points needs to be bound to the specific connection model and linked to specific material and dimensional information to further distinguish the type of key connection points. For example, the type of key connection points corresponding to Q345 steel, M30 anchor bolts, and rear anchor models with specific anchor plate dimensions. The connection state analysis module is used to determine the connection state parameter dataset corresponding to each key connection point based on the correlation between the force response sequence data and the displacement response sequence data of key connection points in the response sequence dataset. The connection state parameter dataset stores connection state parameters, including equivalent stiffness, equivalent connection gap, and hysteresis parameters. The equivalent stiffness represents the ability of the key connection point to resist local deformation after establishing effective force. The equivalent connection gap represents the idle travel or equivalent relaxation amount between the start of relative motion and the establishment of stable force transmission at the key connection point, and is used to reflect pin-hole clearance, fulcrum separation, connection slack, or insufficient preload. The hysteresis parameter represents the degree of non-overlap between the loading and unloading paths of the key connection point, and is used to reflect frictional energy loss, gap closure, local micro-slippage, and irreversible deformation. The anomaly analysis module performs a unified-dimensional transformation on the connection state parameter dataset corresponding to each critical connection point to obtain a standard connection state parameter dataset. It then retrieves a library of anomaly type change trajectories corresponding to the type of the critical connection point. This library includes the change trajectories of all connection state parameters for each anomaly type. The anomaly type corresponding to the critical connection point is determined based on the distribution of the projections from the standard connection state parameter dataset to these change trajectories. Here, the connection state parameters are the change trajectories corresponding to equivalent stiffness, equivalent connection gap, hysteresis parameter, and slip threshold. The anomaly type is set by the operator, such as reduced preload, increased equivalent connection gap, and decreased friction coefficient.

[0017] It should be understood that those skilled in the art can make improvements or modifications based on the above description, and all such improvements and modifications should fall within the protection scope of the appended claims. Parts not described in detail in this specification are prior art known to those skilled in the art.

Claims

1. A method for analyzing the attitude data of a bridge hanging basket, characterized in that, include: Several disturbance action instruction sets are constructed based on linear independence constraints, and all disturbance action instruction sets are applied to the bridge hanging basket in sequence. Response sequence datasets corresponding to all key connection points of the bridge hanging basket are collected. The response sequence datasets include force response sequence data and displacement response sequence data of key connection points. The force response sequence data and displacement response sequence data of key connection points respectively include force response data and displacement response data of key connection points arranged according to the sampling time. Each response sequence dataset is associated with the type of a key connection point. Based on the correlation between the force response sequence data and the displacement response sequence data of the key connection points in the response sequence dataset, the connection state parameter dataset corresponding to each key connection point is determined. The connection state parameter dataset stores the connection state parameters, which include equivalent stiffness, equivalent connection gap and hysteresis parameters. For each critical connection point, after performing a uniform dimension transformation operation on the connection state parameter dataset corresponding to the critical connection point, a standard connection state parameter dataset is obtained. An anomaly type change trajectory library corresponding to the type of the critical connection point is obtained. The anomaly type change trajectory library includes the connection state parameter change trajectory corresponding to all connection state parameters for each anomaly type. The anomaly type corresponding to the critical connection point is determined based on the distribution of the projection of the standard connection state parameter dataset onto the connection state parameter change trajectory.

2. The method for analyzing the attitude data of a bridge hanging basket according to claim 1, characterized in that, Several perturbation action instruction sets are constructed based on linear independence constraints, specifically including the following steps: Obtain the attitude sensitivity matrix, and denote the difference between the number of columns of the attitude sensitivity matrix and the numerical rank of the attitude sensitivity matrix as the null space dimension H. Perform complete singular value decomposition on the attitude sensitivity matrix, and then extract the last H column vectors from the right singular matrix in the output of the singular value decomposition to form the null space basis matrix. When generating K linearly independent candidate perturbation action vectors, an initial coefficient matrix of size H×K is randomly generated. The coefficients in the initial coefficient matrix are random numbers within a set range. The column vectors of the initial coefficient matrix are orthogonalized to obtain a coefficient matrix with full column rank. The null basis matrix is ​​multiplied by the coefficient matrix to obtain a candidate perturbation action matrix of size m×K, where K is not greater than q. Each column in the candidate perturbation action matrix corresponds to a candidate perturbation action vector. The attitude sensitivity matrix is ​​constructed as follows: several bridge cantilever formwork attitude analysis samples are obtained, including disturbance motion vectors and corresponding bridge cantilever formwork attitude vectors. The disturbance motion vectors include control quantities corresponding to several actuators. The attitude analysis model is constructed by multiplying the attitude sensitivity matrix and the transpose of the disturbance motion vectors, which equals the transpose of the bridge cantilever formwork attitude vectors. Then, all bridge cantilever formwork attitude analysis samples are substituted into the attitude analysis model, and a regression operation is performed on the attitude sensitivity matrix.

3. The method for analyzing the attitude data of a bridge hanging basket according to claim 2, characterized in that, The connection state parameter dataset corresponding to each key connection point is determined based on the correlation between the force response sequence data and the displacement response sequence data of key connection points in the response sequence dataset. The specific steps include the following: The force response data and displacement response data of the key connection points at the same sampling time are time-aligned, and the force response data and displacement response data at the first sampling time in the force response sequence data and displacement response sequence data of the key connection points are subtracted respectively to obtain the force increment data and the local relative displacement increment data; the force increment data and the local relative displacement increment data at the same time are combined into response data points according to the sampling time order. Using local relative displacement increment data as independent variables and force increment data as dependent variables, linear regression is performed on all response data points to construct a response linear relationship. The slope in the response linear relationship is determined as the equivalent stiffness corresponding to the key connection point. The local relative displacement increment data corresponding to the force increment data of 0 in the linear response relationship is determined as the equivalent connection gap corresponding to the key connection point. The trajectory area formed between adjacent response data points is calculated according to the sampling time sequence, and the trajectory area is normalized by the variation range of the force increment data and the variation range of the local relative displacement increment data to obtain the hysteresis parameter.

4. The method for analyzing the attitude data of a bridge hanging basket according to claim 3, characterized in that, The anomaly type change trajectory library is constructed through the following steps: For any critical connection point, determine the physical degradation variable corresponding to the type of the anomaly, obtain several anomaly analysis samples, including the physical degradation variable and the corresponding standard connection state parameter dataset, and combine all the anomaly analysis samples into an anomaly analysis sample set corresponding to the anomaly type of the critical connection point. For any critical join point, the connection state parameters in the connection state parameter dataset are traversed, and the currently selected connection state parameter is determined as the target connection state parameter. The physical degradation variables and target connection state parameters contained in each anomaly analysis sample are extracted from the anomaly analysis sample set corresponding to the anomaly type. These two are combined to form a fitting point set, with the physical degradation variables as independent variables and the target connection state parameters as dependent variables. The sample fitting error is the sum of the squares of the differences between the predicted values ​​of the intermediate fitting function at each physical degradation variable and the corresponding target connection state parameter. The second derivative of the intermediate fitting function within the interval corresponding to the physical degradation variables covered by all anomaly analysis samples in the anomaly analysis sample set is used as the fitting error. The integral is used to obtain a smoothing penalty term. The sum of the product of the sample fitting error, the smoothing penalty term, and the smoothing coefficient is then used as the fitting objective. By adjusting the coefficients of the spline function in the intermediate fitting function, the fitting objective is minimized, resulting in the trajectory of the connection state parameters corresponding to the target connection state parameters. Each trajectory of the connection state parameters is a continuous function between the physical degradation variable and the corresponding connection state parameter. This process continues until all connection state parameters in the connection state parameter dataset have been traversed. All the trajectory of the connection state parameters are then combined into a set of anomaly type trajectory corresponding to the anomaly type. Finally, the sets of anomaly type trajectory corresponding to all anomaly types under the type of the key connection point are combined into an anomaly type trajectory library.

5. The method for analyzing the attitude data of a bridge hanging basket according to claim 4, characterized in that, The anomaly type corresponding to the critical connection point is determined based on the distribution of the projection from the standard connection state parameter dataset to the connection state parameter change trajectory. The specific steps include the following: Obtain the anomaly type change trajectory library, traverse all anomaly types, and for the selected anomaly type, record the square of the difference between the standard connection state parameter and the connection state parameter change trajectory corresponding to the target connection state parameter corresponding to the standard connection state parameter as the connection state parameter difference function. Then, perform a weighted summation operation on all connection state parameter difference functions to obtain the joint interpretation error function. The anomaly type corresponding to the critical join point is output if the optimal matching value is less than the matching threshold and the physical degradation variable corresponding to the optimal matching value is higher than the degradation threshold. If there is no anomaly type where the optimal matching value is less than the matching threshold and the physical degradation variable corresponding to the optimal matching value is higher than the degradation threshold, then the critical join point does not have anomalies.

6. The method for analyzing the attitude data of a bridge hanging basket according to claim 5, characterized in that, After performing a uniform dimensional transformation on the connection state parameter dataset corresponding to the key connection points, a standard connection state parameter dataset is obtained. The specific steps include the following: For the connection state parameters in the connection state parameter dataset, the ratio of the connection state parameter to the corresponding baseline quantity is recorded as the standard connection state parameter. Then, all the standard connection state parameters are combined into a standard connection state parameter dataset.

7. A bridge hanging basket attitude data analysis system, characterized in that, The system employs a bridge hanging basket attitude data analysis method according to any one of claims 1-6, comprising: The disturbance action instruction set construction module is used to construct several disturbance action instruction sets based on linearly independent constraints, and apply all disturbance action instruction sets to the bridge hanging basket in sequence. It also collects response sequence datasets corresponding to all key connection points of the bridge hanging basket. The response sequence datasets include key connection point force response sequence data and key connection point displacement response sequence data. The key connection point force response sequence data and key connection point displacement response sequence data respectively include key connection point force response data and key connection point displacement response data arranged according to the sampling time. Each response sequence dataset is associated with the type of a key connection point. The connection state analysis module is used to determine the connection state parameter dataset corresponding to each key connection point based on the correlation between the force response sequence data and the displacement response sequence data of the key connection points in the response sequence dataset. The connection state parameter dataset stores the connection state parameters, which include equivalent stiffness, equivalent connection gap and hysteresis parameters. The anomaly analysis module is used to perform a uniform dimension transformation operation on the connection state parameter dataset corresponding to each key connection point to obtain a standard connection state parameter dataset. It then obtains a library of anomaly type change trajectories corresponding to the type of the key connection point. The library of anomaly type change trajectories includes the connection state parameter change trajectories corresponding to all connection state parameters for each anomaly type. Based on the distribution of the projection of the standard connection state parameter dataset onto the connection state parameter change trajectories, the anomaly type corresponding to the key connection point is determined.

Images